[c-nsp] Strange behaviour with 6509

Church, Chuck cchurch at netcogov.com
Fri Feb 10 10:53:11 EST 2006

What's the round trip time in ms between your machine and the FTP server?  Sounds a lot like your RTT is limiting your TCP session due to windowing and waiting for ACKs. 

Chuck Church
Network Engineer
Netco Government Services -  Enterprise Network Engineering
1210 N. Parker Rd.
Greenville, SC 29609
Home office: 864-335-9473
Cell: 864-266-3978
cchurch at netcogov.com
PGP key: http://pgp.mit.edu:11371/pks/lookup?op=get&search=0x4371A48D
"I'm one Snickers Pie away from losing my foot to diabetes."  -  Homer

-----Original Message-----
From: cisco-nsp-bounces at puck.nether.net [mailto:cisco-nsp-bounces at puck.nether.net] On Behalf Of Ahmad Cheikh Moussa
Sent: Friday, February 10, 2006 5:20 AM
To: cisco-nsp at puck.nether.net
Subject: [c-nsp] Strange behaviour with 6509


I have a 6509 with Sup 2 and a lot of WS-X6416-GBIC modules.
So I have a strange behaviour. If I make a ftp test
connection through  the catalyst, I only got a throughput
of approximately 1.4Mbyte/s. Now, if the catalyst has a
bottleneck, and I only have 1.4Mbyte/s then this bandwith
should be shared for all other connections. So I made
a test with 10 paralel connections from one machine.
Normally, the 10 connections should share this bandwith,
but I got for every connection 1.4Mbyte/s. That means,
I have approximatley 14 Mbyte/s.

So why I don't get 14 Mbyte/s, when I only use one
connection ? I don't understand this behaviour.

Can someone give me a hint ?
Could this be a cef bug ?

Thanks in advance,

P.S.: On the 6509 I can not disable the cef switching.
Ahmad Cheikh-Moussa
Dr.-Hell-Straße, 24107 Kiel, Germany
Telefon: +49 431 2390 400 --  Telefax: +49 431 2390 499
Service: Service at NetUSE.DE --  http://NetUSE.DE/

cisco-nsp mailing list  cisco-nsp at puck.nether.net
archive at http://puck.nether.net/pipermail/cisco-nsp/

More information about the cisco-nsp mailing list